Attempts to devise a serodiagnostic test for TB have been made for decades with disappointing results, primarily because most of the antigens evaluated are not relevant to the human immune response. During the last 7 years, we have characterized the humoral immune responses in TB patients at different stages of active disease. These studies form the basis for the research that is now needed to develop a serodiagnostic test for TB. Our studies clearly show that antigens chosen on the basis of their immunodominance during different stages of active TB, and in both HIV-infected and non-HIV TB patients, will provide the greatest opportunity to develop a successful diagnostic test. In this context, we have already identified approximately 12 proteins in culture filtrates of M. tuberculosis that elicit antibodies in patients with incipient pre-clinical TB, non-cavitary TB or cavitary TB. This subset of antigens also elicits antibodies in patients co-infected with HIV and TB. Two proteins of this subset have been cloned, and provide a serodiagnostic assay with the highest sensitivity and specificity that has yet been achieved with any antigens. In the current application, we propose to: (a) Identify the additional immunodominant antigens. This will be done both by proteomic and molecular approaches; (b) Clone the genes for these antigens into E.coli and evaluate the reactivity of the recombinant proteins with antibodies from TB patients at different stages of TB; (c) Map the immunodominant epitopes on these candidate proteins by a variety of strategies; (d) Identify the immunodominant peptides of these antigens that are recognized by antibodies from patients across the spectrum of TB; and finally (e) Select and evaluate the combinations of peptides that will be the basis of a low-cost, rapid, point-of-care diagnostic test for TB with high sensitivity and specificity.